Upload folder using huggingface_hub
Browse files- config.json +46 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +368 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"id2label": {
|
| 15 |
+
"0": "fear",
|
| 16 |
+
"1": "anger",
|
| 17 |
+
"2": "sadness",
|
| 18 |
+
"3": "disgust",
|
| 19 |
+
"4": "joy",
|
| 20 |
+
"5": "surprise"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 3072,
|
| 24 |
+
"is_decoder": false,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"anger": 1,
|
| 27 |
+
"disgust": 3,
|
| 28 |
+
"fear": 0,
|
| 29 |
+
"joy": 4,
|
| 30 |
+
"sadness": 2,
|
| 31 |
+
"surprise": 5
|
| 32 |
+
},
|
| 33 |
+
"layer_norm_eps": 1e-05,
|
| 34 |
+
"max_position_embeddings": 514,
|
| 35 |
+
"model_type": "xlm-roberta",
|
| 36 |
+
"num_attention_heads": 12,
|
| 37 |
+
"num_hidden_layers": 12,
|
| 38 |
+
"output_past": true,
|
| 39 |
+
"pad_token_id": 1,
|
| 40 |
+
"position_embedding_type": "absolute",
|
| 41 |
+
"tie_word_embeddings": true,
|
| 42 |
+
"transformers_version": "5.10.1",
|
| 43 |
+
"type_vocab_size": 1,
|
| 44 |
+
"use_cache": false,
|
| 45 |
+
"vocab_size": 250002
|
| 46 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8fd230e365c0fbbae1ddaf749b2b0655ac5b29f503a5f2d1960d46cd32ea3c8
|
| 3 |
+
size 1112217312
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2edef80e169f724aa06e88928dc3f1254c8f1c8122caa1c31e5a25a9cb548453
|
| 3 |
+
size 2224557451
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6aac33d382868a168809d2c3ac8c7efb276c1f597e376a9a61c2cba28881592a
|
| 3 |
+
size 14645
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:040c31ecf895d199328434c37d5d66f5b653c781b0d4d25db730ae83754eeceb
|
| 3 |
+
size 1383
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7b2c5db660e6409b97c61c496365a053bb14ef6692f6a348b697690d68ec5fd
|
| 3 |
+
size 1465
|
trainer_state.json
ADDED
|
@@ -0,0 +1,368 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 20025,
|
| 3 |
+
"best_metric": 0.6554931335830212,
|
| 4 |
+
"best_model_checkpoint": "/content/drive/MyDrive/final/roberta_emotion_model/checkpoint-20025",
|
| 5 |
+
"epoch": 5.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 20025,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.12484394506866417,
|
| 14 |
+
"grad_norm": 48.22937774658203,
|
| 15 |
+
"learning_rate": 9.750811485642947e-06,
|
| 16 |
+
"loss": 3.1020751953125,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.24968789013732834,
|
| 21 |
+
"grad_norm": 28.517393112182617,
|
| 22 |
+
"learning_rate": 9.50112359550562e-06,
|
| 23 |
+
"loss": 2.5971416015625,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.37453183520599254,
|
| 28 |
+
"grad_norm": 45.27549743652344,
|
| 29 |
+
"learning_rate": 9.25143570536829e-06,
|
| 30 |
+
"loss": 2.350372802734375,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.4993757802746567,
|
| 35 |
+
"grad_norm": 63.087188720703125,
|
| 36 |
+
"learning_rate": 9.001747815230962e-06,
|
| 37 |
+
"loss": 2.30302587890625,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.6242197253433208,
|
| 42 |
+
"grad_norm": 68.39694213867188,
|
| 43 |
+
"learning_rate": 8.752059925093633e-06,
|
| 44 |
+
"loss": 2.235894287109375,
|
| 45 |
+
"step": 2500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.7490636704119851,
|
| 49 |
+
"grad_norm": 48.36997985839844,
|
| 50 |
+
"learning_rate": 8.502372034956306e-06,
|
| 51 |
+
"loss": 2.18744384765625,
|
| 52 |
+
"step": 3000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.8739076154806492,
|
| 56 |
+
"grad_norm": 63.88243865966797,
|
| 57 |
+
"learning_rate": 8.252684144818976e-06,
|
| 58 |
+
"loss": 2.1574775390625,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.9987515605493134,
|
| 63 |
+
"grad_norm": 25.953285217285156,
|
| 64 |
+
"learning_rate": 8.002996254681649e-06,
|
| 65 |
+
"loss": 2.149793701171875,
|
| 66 |
+
"step": 4000
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.0,
|
| 70 |
+
"eval_accuracy": 0.5983770287141074,
|
| 71 |
+
"eval_loss": 1.0222212076187134,
|
| 72 |
+
"eval_runtime": 31.588,
|
| 73 |
+
"eval_samples_per_second": 507.154,
|
| 74 |
+
"eval_steps_per_second": 31.721,
|
| 75 |
+
"step": 4005
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 1.1235955056179776,
|
| 79 |
+
"grad_norm": 65.45611572265625,
|
| 80 |
+
"learning_rate": 7.753308364544321e-06,
|
| 81 |
+
"loss": 1.9485059814453125,
|
| 82 |
+
"step": 4500
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 1.2484394506866416,
|
| 86 |
+
"grad_norm": 79.43843078613281,
|
| 87 |
+
"learning_rate": 7.503620474406992e-06,
|
| 88 |
+
"loss": 1.950569580078125,
|
| 89 |
+
"step": 5000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.373283395755306,
|
| 93 |
+
"grad_norm": 38.40707015991211,
|
| 94 |
+
"learning_rate": 7.253932584269664e-06,
|
| 95 |
+
"loss": 1.957847900390625,
|
| 96 |
+
"step": 5500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.4981273408239701,
|
| 100 |
+
"grad_norm": 29.66714096069336,
|
| 101 |
+
"learning_rate": 7.004244694132335e-06,
|
| 102 |
+
"loss": 1.90708447265625,
|
| 103 |
+
"step": 6000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.6229712858926342,
|
| 107 |
+
"grad_norm": 28.491024017333984,
|
| 108 |
+
"learning_rate": 6.754556803995007e-06,
|
| 109 |
+
"loss": 1.911080078125,
|
| 110 |
+
"step": 6500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.7478152309612984,
|
| 114 |
+
"grad_norm": 94.18211364746094,
|
| 115 |
+
"learning_rate": 6.504868913857678e-06,
|
| 116 |
+
"loss": 1.8992110595703124,
|
| 117 |
+
"step": 7000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.8726591760299627,
|
| 121 |
+
"grad_norm": 28.84901237487793,
|
| 122 |
+
"learning_rate": 6.25518102372035e-06,
|
| 123 |
+
"loss": 1.887749755859375,
|
| 124 |
+
"step": 7500
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.9975031210986267,
|
| 128 |
+
"grad_norm": 51.08824920654297,
|
| 129 |
+
"learning_rate": 6.005493133583021e-06,
|
| 130 |
+
"loss": 1.891501220703125,
|
| 131 |
+
"step": 8000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.0,
|
| 135 |
+
"eval_accuracy": 0.6292759051186018,
|
| 136 |
+
"eval_loss": 0.942175567150116,
|
| 137 |
+
"eval_runtime": 31.7611,
|
| 138 |
+
"eval_samples_per_second": 504.391,
|
| 139 |
+
"eval_steps_per_second": 31.548,
|
| 140 |
+
"step": 8010
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.1223470661672907,
|
| 144 |
+
"grad_norm": 41.6335563659668,
|
| 145 |
+
"learning_rate": 5.755805243445693e-06,
|
| 146 |
+
"loss": 1.7376878662109374,
|
| 147 |
+
"step": 8500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.247191011235955,
|
| 151 |
+
"grad_norm": 54.421661376953125,
|
| 152 |
+
"learning_rate": 5.506117353308364e-06,
|
| 153 |
+
"loss": 1.723041748046875,
|
| 154 |
+
"step": 9000
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 2.3720349563046192,
|
| 158 |
+
"grad_norm": 39.90679168701172,
|
| 159 |
+
"learning_rate": 5.256429463171036e-06,
|
| 160 |
+
"loss": 1.71384814453125,
|
| 161 |
+
"step": 9500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 2.4968789013732833,
|
| 165 |
+
"grad_norm": 35.013450622558594,
|
| 166 |
+
"learning_rate": 5.006741573033709e-06,
|
| 167 |
+
"loss": 1.6872838134765624,
|
| 168 |
+
"step": 10000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 2.6217228464419478,
|
| 172 |
+
"grad_norm": 26.661258697509766,
|
| 173 |
+
"learning_rate": 4.7570536828963795e-06,
|
| 174 |
+
"loss": 1.6985811767578125,
|
| 175 |
+
"step": 10500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.746566791510612,
|
| 179 |
+
"grad_norm": 30.96997833251953,
|
| 180 |
+
"learning_rate": 4.5073657927590515e-06,
|
| 181 |
+
"loss": 1.66519970703125,
|
| 182 |
+
"step": 11000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.871410736579276,
|
| 186 |
+
"grad_norm": 32.575050354003906,
|
| 187 |
+
"learning_rate": 4.257677902621723e-06,
|
| 188 |
+
"loss": 1.6637918701171874,
|
| 189 |
+
"step": 11500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.9962546816479403,
|
| 193 |
+
"grad_norm": 44.792823791503906,
|
| 194 |
+
"learning_rate": 4.007990012484395e-06,
|
| 195 |
+
"loss": 1.6791025390625,
|
| 196 |
+
"step": 12000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 3.0,
|
| 200 |
+
"eval_accuracy": 0.6441323345817728,
|
| 201 |
+
"eval_loss": 0.9494165182113647,
|
| 202 |
+
"eval_runtime": 31.7182,
|
| 203 |
+
"eval_samples_per_second": 505.072,
|
| 204 |
+
"eval_steps_per_second": 31.591,
|
| 205 |
+
"step": 12015
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 3.1210986267166043,
|
| 209 |
+
"grad_norm": 66.56997680664062,
|
| 210 |
+
"learning_rate": 3.7583021223470666e-06,
|
| 211 |
+
"loss": 1.5335745849609375,
|
| 212 |
+
"step": 12500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.2459425717852683,
|
| 216 |
+
"grad_norm": 63.02039337158203,
|
| 217 |
+
"learning_rate": 3.508614232209738e-06,
|
| 218 |
+
"loss": 1.51216015625,
|
| 219 |
+
"step": 13000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 3.370786516853933,
|
| 223 |
+
"grad_norm": 60.7429313659668,
|
| 224 |
+
"learning_rate": 3.2589263420724098e-06,
|
| 225 |
+
"loss": 1.5589078369140625,
|
| 226 |
+
"step": 13500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 3.495630461922597,
|
| 230 |
+
"grad_norm": 33.59284210205078,
|
| 231 |
+
"learning_rate": 3.0092384519350813e-06,
|
| 232 |
+
"loss": 1.554709228515625,
|
| 233 |
+
"step": 14000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.620474406991261,
|
| 237 |
+
"grad_norm": 83.69245910644531,
|
| 238 |
+
"learning_rate": 2.759550561797753e-06,
|
| 239 |
+
"loss": 1.49451025390625,
|
| 240 |
+
"step": 14500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.7453183520599254,
|
| 244 |
+
"grad_norm": 58.14004898071289,
|
| 245 |
+
"learning_rate": 2.5098626716604245e-06,
|
| 246 |
+
"loss": 1.511951904296875,
|
| 247 |
+
"step": 15000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 3.8701622971285894,
|
| 251 |
+
"grad_norm": 47.35211944580078,
|
| 252 |
+
"learning_rate": 2.260174781523096e-06,
|
| 253 |
+
"loss": 1.5268017578125,
|
| 254 |
+
"step": 15500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.9950062421972534,
|
| 258 |
+
"grad_norm": 64.91954040527344,
|
| 259 |
+
"learning_rate": 2.010486891385768e-06,
|
| 260 |
+
"loss": 1.512477294921875,
|
| 261 |
+
"step": 16000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.0,
|
| 265 |
+
"eval_accuracy": 0.6526841448189763,
|
| 266 |
+
"eval_loss": 0.9551559090614319,
|
| 267 |
+
"eval_runtime": 31.4775,
|
| 268 |
+
"eval_samples_per_second": 508.935,
|
| 269 |
+
"eval_steps_per_second": 31.832,
|
| 270 |
+
"step": 16020
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 4.119850187265918,
|
| 274 |
+
"grad_norm": 54.49357604980469,
|
| 275 |
+
"learning_rate": 1.7607990012484397e-06,
|
| 276 |
+
"loss": 1.39636083984375,
|
| 277 |
+
"step": 16500
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 4.2446941323345815,
|
| 281 |
+
"grad_norm": 91.57239532470703,
|
| 282 |
+
"learning_rate": 1.5111111111111112e-06,
|
| 283 |
+
"loss": 1.3841357421875,
|
| 284 |
+
"step": 17000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.369538077403246,
|
| 288 |
+
"grad_norm": 103.39310455322266,
|
| 289 |
+
"learning_rate": 1.2614232209737828e-06,
|
| 290 |
+
"loss": 1.3866033935546875,
|
| 291 |
+
"step": 17500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.49438202247191,
|
| 295 |
+
"grad_norm": 68.51345825195312,
|
| 296 |
+
"learning_rate": 1.0117353308364546e-06,
|
| 297 |
+
"loss": 1.3865987548828125,
|
| 298 |
+
"step": 18000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 4.619225967540574,
|
| 302 |
+
"grad_norm": 61.58140563964844,
|
| 303 |
+
"learning_rate": 7.620474406991262e-07,
|
| 304 |
+
"loss": 1.413747802734375,
|
| 305 |
+
"step": 18500
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 4.7440699126092385,
|
| 309 |
+
"grad_norm": 24.058151245117188,
|
| 310 |
+
"learning_rate": 5.123595505617978e-07,
|
| 311 |
+
"loss": 1.393927978515625,
|
| 312 |
+
"step": 19000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 4.868913857677903,
|
| 316 |
+
"grad_norm": 39.903724670410156,
|
| 317 |
+
"learning_rate": 2.6267166042446943e-07,
|
| 318 |
+
"loss": 1.436919677734375,
|
| 319 |
+
"step": 19500
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 4.9937578027465666,
|
| 323 |
+
"grad_norm": 59.79766845703125,
|
| 324 |
+
"learning_rate": 1.2983770287141074e-08,
|
| 325 |
+
"loss": 1.3762628173828124,
|
| 326 |
+
"step": 20000
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 5.0,
|
| 330 |
+
"eval_accuracy": 0.6554931335830212,
|
| 331 |
+
"eval_loss": 0.9726009964942932,
|
| 332 |
+
"eval_runtime": 31.4744,
|
| 333 |
+
"eval_samples_per_second": 508.985,
|
| 334 |
+
"eval_steps_per_second": 31.835,
|
| 335 |
+
"step": 20025
|
| 336 |
+
}
|
| 337 |
+
],
|
| 338 |
+
"logging_steps": 500,
|
| 339 |
+
"max_steps": 20025,
|
| 340 |
+
"num_input_tokens_seen": 0,
|
| 341 |
+
"num_train_epochs": 5,
|
| 342 |
+
"save_steps": 500,
|
| 343 |
+
"stateful_callbacks": {
|
| 344 |
+
"EarlyStoppingCallback": {
|
| 345 |
+
"args": {
|
| 346 |
+
"early_stopping_patience": 2,
|
| 347 |
+
"early_stopping_threshold": 0.0
|
| 348 |
+
},
|
| 349 |
+
"attributes": {
|
| 350 |
+
"early_stopping_patience_counter": 0
|
| 351 |
+
}
|
| 352 |
+
},
|
| 353 |
+
"TrainerControl": {
|
| 354 |
+
"args": {
|
| 355 |
+
"should_epoch_stop": false,
|
| 356 |
+
"should_evaluate": false,
|
| 357 |
+
"should_log": false,
|
| 358 |
+
"should_save": true,
|
| 359 |
+
"should_training_stop": true
|
| 360 |
+
},
|
| 361 |
+
"attributes": {}
|
| 362 |
+
}
|
| 363 |
+
},
|
| 364 |
+
"total_flos": 2.107529463578112e+16,
|
| 365 |
+
"train_batch_size": 8,
|
| 366 |
+
"trial_name": null,
|
| 367 |
+
"trial_params": null
|
| 368 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe091723f23ed1c5f6ce73d481555956a4931eefeb0b0b1c41e198e36a79d15c
|
| 3 |
+
size 5201
|